Of all enzymes, triosephosphate isomerase is one of the most completely defined in both kinetic and structural terms. The complete free energy profile was established in our laboratory in 1976, and the crystal structures of isomerases from both chicken muscle and from yeast have been solved to high resolution. Structures of complexes of the enzyme with both substrate and inhibitors are also available. From this firm basis of structure:function correlation, we aim to use both site-directed mutagenesis and classical genetic selection to define the nature of the three catalytic elements (the basic residue Glu-165, the acidic residues His-95 and Lys-13, and the 'loop' that closes on bound substrate) and delineate their contribution to the catalytic act. We shall evaluate the complete energetics for each mutant that we generate, and we aim (in collaboration with the crystallographers, and by NMR, FTIR, etc.) to define the precise structures of these mutants. Finally, starting with mutants that we have already produced, the activity of which has been severely reduced, we shall select for more active second site suppressor mutations, in order to illuminate the evolutionary path forward and to test, for the first time, the theories of the development of the catalytic function of enzymes.